Reconsidering SMT Over NMT for Closely Related Languages: A Case Study of Persian-Hindi Pair

TL;DR

This study shows that for closely related languages like Persian-Hindi, traditional Phrase-Based SMT can outperform Transformer NMT in moderate-resource settings, emphasizing the importance of language-specific architecture choices.

Contribution

The paper provides a comparative analysis demonstrating SMT's superior performance over NMT for structurally similar languages in moderate-resource scenarios, with innovative variations like Romanization and sentence reordering.

Findings

PBSMT achieves BLEU of 66.32, surpassing Transformer NMT's 53.7

Variations like Romanized training improve translation quality

SMT remains competitive for closely related languages in moderate-resource contexts

Abstract

This paper demonstrates that Phrase-Based Statistical Machine Translation (PBSMT) can outperform Transformer-based Neural Machine Translation (NMT) in moderate-resource scenarios, specifically for structurally similar languages, like the Persian-Hindi pair. Despite the Transformer architecture's typical preference for large parallel corpora, our results show that PBSMT achieves a BLEU score of 66.32, significantly exceeding the Transformer-NMT score of 53.7 on the same dataset. Additionally, we explore variations of the SMT architecture, including training on Romanized text and modifying the word order of Persian sentences to match the left-to-right (LTR) structure of Hindi. Our findings highlight the importance of choosing the right architecture based on language pair characteristics and advocate for SMT as a high-performing alternative, even in contexts commonly dominated by NMT.